|Year : 2021 | Volume
| Issue : 1 | Page : 12-18
Comparison of setup variations and defining a CTV-PTV margin in fractionated radiotherapy of brain tumor using Novalis Tx Orthogonal OBI (kv) versus Brain Lab ExacTrac (kv) imaging systems with brain lab mask
Hemant K Pandey1, Mano Bhadauria1, Ganesh Kishan Rao Jadhav1, Sapna Manocha1, Renuka Masodkar1, Divya Piyushi1, Sanjay Raut2, Brijesh Goswami2, Sunil Kumar Chauhan3
1 Department of Radiation Oncology, Indraprastha Apollo Hospital, New Delhi, India
2 Department of Medical Physicist, Indraprastha Apollo Hospital, New Delhi, India
3 Shobhit Institute of Engineering and Technology (Deemed to be University), Meerut, Uttar Pradesh, India
|Date of Submission||12-Sep-2021|
|Date of Acceptance||29-Nov-2021|
|Date of Web Publication||06-Jan-2022|
Dr. Hemant K Pandey
Department of Radiation Oncology, Indraprastha Apollo Hospital, 668, Pocket A, Sarita Vihar, New Delhi 110076.
Source of Support: None, Conflict of Interest: None
The objective of this study is to evaluate the setup discrepancy between Brain Lab ExacTrac (kv) and Orthogonal Novalis Tx (kv) imaging systems and to obtain an optimal CTV-PTV margin. We recruited 35 consecutive brain tumor patients, immobilized with non-invasive thermoplastic BrainLab mask, between August 2016 and August 2017. The position verification done by ExacTrac, Infrared Positioning System was followed by a set of images by OBI at 0° and 90° position. The images were autofused and compared with DRR generated by TPS. The extent of translational and rotational shift was then applied and documented. The translational inter-fractional setup error (mean ± SD) in X-axis detected by OBI and ET was 0.002 ± 0.227 and −0.042 ± 0.084, respectively. In Y-axis, the error detected was −0.01 ± 0.176 and 0.075 ± 0.079, respectively, whereas in Z-axis the error with OBI and ET was 0.329 ± 0.521 and 0.051 ± 0.091, respectively. The P-value of inter-fractional errors in X, Y, and Z axes was 0.294, 0.013, and 0.004, which is statistically significant in Y- and Z-axes. Rotational inter-fractional setup errors (mean ± SD) for OBI and ET were −0.101 ± 0.257 and −0.076 ± 0.102, respectively. The rotational setup errors detected by ET in pitch and roll dimension are −0.003 ± 0.084 and −0.085 ± 0.073, respectively. The data obtained were put in Van Herk’s formula (PTV margin = 2.5 ∑ + 0.74 σ), and CTV-PTV margin was calculated. CTV-PTV margin for ET was 2.70, 2.72, and 2.94 mm and 3.38o, 2.84o, and 2.51o in lateral, longitudinal, vertical and pitch, roll, yaw dimension, respectively. The CTV-PTV margin for OBI was 7.40, 6.20, and 5.30 mm and 7.27o in lateral, longitudinal, vertical, and yaw dimension, respectively. Daily setup verifications that use the Brain LAB Exac Trac 6D image-guided system are very useful for brain tumor treatment; nevertheless, OBI kv imaging can be used for FRT of brain tumor with extended margin.
Keywords: Brain Lab, Exac Trac, OBI, random errors, systemic errors, von Herk
|How to cite this article:|
Pandey HK, Bhadauria M, Jadhav GK, Manocha S, Masodkar R, Piyushi D, Raut S, Goswami B, Chauhan SK. Comparison of setup variations and defining a CTV-PTV margin in fractionated radiotherapy of brain tumor using Novalis Tx Orthogonal OBI (kv) versus Brain Lab ExacTrac (kv) imaging systems with brain lab mask. Bengal J Cancer 2021;1:12-8
|How to cite this URL:|
Pandey HK, Bhadauria M, Jadhav GK, Manocha S, Masodkar R, Piyushi D, Raut S, Goswami B, Chauhan SK. Comparison of setup variations and defining a CTV-PTV margin in fractionated radiotherapy of brain tumor using Novalis Tx Orthogonal OBI (kv) versus Brain Lab ExacTrac (kv) imaging systems with brain lab mask. Bengal J Cancer [serial online] 2021 [cited 2023 Feb 8];1:12-8. Available from: http://www.bengaljcancer.org/text.asp?2021/1/1/12/335059
| Introduction|| |
The underlying goal of treating brain tumors with radiation is to maximize the therapeutic ratio by delivering high dose to the tumor while restricting the dose to the normal brain structures. This in turn is possible by use of modern radiation techniques such as intensity modulated radiotherapy (IMRT) and image-guided radiotherapy (IGRT). Image guidance aids in precision while delivering radiation to the target and helps in reduction of setup uncertainty.
IGRT plays an important role in the treatment of brain tumors. A number of imaging modalities, including two-dimensional radio imaging (X-ray), computed tomography (CT-scan), orthogonal on board imaging (OBI-kv), cone beam computed tomography (CBCT), electronic portal imaging device (EPID), ultrasonography, PET scan, and magnetic resonance imaging (MRI), are used in IGRT to help in accurate target positioning and treatment delivery.,
The radiotherapy process includes many uncertainties that should be considered in the planning target volume definition in order to avoid any geometrical lack during treatment delivery. Setup errors should be considered. Patient setup error can be defined as the difference between the actual and the planned position of the patient with respect to the treatment beams during irradiation. The setup error can be of two types, systematic error (∑) and random error (α). Systematic error (∑) can be defined as a patient-dependent average displacement between treatment anatomy and planning scan anatomy. Random error (α), in contrast, is the day-to-day deviation from the average target position, introduced mainly due to internal organ motion, breathing, or it can be patient-related (weight loss or gain, sedative drugs used during treatment). The distinction between a systematic and a random component reflects a different impact on treatment dose. In fact, it has been demonstrated that to ensure dose homogeneity coverage for target volume, systematic errors require margins three to four times as large as comparable random displacements.,
In this study, we have compared correction of setup variation with respect to Linac Orthogonal On-Board Imaging (OBI-kv) versus Orthogonal Brain Lab ExacTrac-KV (ET-kv) imaging modalities (Brain-Lab, Feldkirchen, Germany) using Brain Lab Mask for brain tumor patients. However, both imaging modalities are two-dimensional planar X-ray imaging system and use less information for image registration, in comparison to CBCT, which is a three-dimensional volumetric imaging. Thus it would be interesting to compare image registration obtained by orthogonal OBI-kv and orthogonal Brain Lab ET-kv imaging systems.
The OBI-kv system is a tool for IGRT which is mounted on a Novalis Tx, linear accelerator. The combined Linac OBI-kv system provides diagnostic quality images of patient’s internal anatomy with high positional accuracy and high contrast. It is a four-dimensional (4DOF) imaging system that can correct setup errors in all translational (lateral, vertical, and longitudinal) and one rotational (yaw) dimension, leaving some uncertainty in other two rotational (pitch and roll) dimensions.
The Brain Lab ExacTrac imaging system has six degrees of freedom, and thus it can correct setup errors in all translational and rotational dimensions. It is mainly an integration of two subsystems: an infrared (IR)-based optical positioning system for initial patient setup and precise control of couch movement, and an orthogonal (kv) X-ray imaging system with two X-ray sources located on the floor and two detectors mounted on the ceiling is free of couch collision, for position verification and adjustment. The whole system is integrated with a linear accelerator along with a robotic couch with six degrees of freedom.
The purpose of this study was to compare the two modalities of image guidance in respect of finding extent of setup errors and correction of those errors and to define a PTV margin with safe therapeutic index, from the data extracted in the treatment of brain tumors. If the PTV margin obtained by OBI is comparable to that obtained by ET, it may be used as IGRT for brain tumor treatment in centers where ET is not available.
| Materials and Methods|| |
This is a prospective comparative study between two imaging modalities, orthogonal OBI-kv and BrainLab ExacTrac-kv, for setup correction and verification before delivering radiotherapy. The ExacTrac was used as control arm and OBI as study arm. The study was conducted and patients were recruited from August 2016 to August 2017 at Radiotherapy Department, Indraprastha Apollo Hospital, New Delhi.
Novalis-Tx (Varian Medical Systems, USA), linear accelerator integrated with Brain Lab ExacTrac (Brain Lab, Feldkirchen, Germany, version 6.1.1) imaging and positioning system, and orthogonal OBI system and a robotic couch (Robotic Couch®, BrainLab, Germany) with six-degree-of-freedom (DOF) positioning and correction system were used for this study. Each patient received a total of 25–30 fractions with 6 MV beam energy. Correction and verification images were acquired on alternate day of treatment. A total of 21,00 images were performed for the 35 patients on ExacTrac system and 1,050 images on OBI system, and the data obtained in translational and rotational dimensions were used for analysis.
The patients were immobilized with a non-invasive thermoplastic BrainLab mask (Orfit, Thermoplastic Mask by Brain Lab AG). It has three different layers. One layer is moulded to the back of the patient’s head. A second layer contains three reinforcing straps, on forehead, below nose, and over the chin. A third layer is placed over the mask with three straps. These masks are attached to the couch-mounted support system which provides rigid fixation of the patient’s head. Simulation CT scans were done with patient’s head with brain lab mask placed in a localizer box. The localizer box provides the stereotactic reference coordinates, which help in determining the localization of the brain tumor.
The Brain Lab, I-plan (version 3.0) treatment planning systems (TPS) was used to define a reproducible and precise stereotactic reference coordinate using a Brain Lab localizer box. One-to-two-millimeter-thick CT scan slices were taken for radiotherapy planning. All CT scan slices were then localized and fused with MRI scans (T1, T1c, T2/Flair). Subsequently, GTV, CTV, PTV, and OARs were delineated using RTOG brain contouring guidelines. All patients received a dose of 50–60 Gy with once-daily fraction at 1.8–2.0 Gy per fraction over 5–6 weeks.
All patients were treated in supine position. After positioning along iso-center using infrared camera on robotic couch, two radiographic images (X-ray correction, XC) were acquired and matched with reference DRRs generated by TPS.
Matching of DRRs and radiographic images is performed using bony anatomy. This fusion generates translational and rotational 6D couch shift. After application of a couch shift, a second set of radiograph images (X-ray verification, XV) were acquired to verify the patient’s position based on bony anatomy and to check whether final translational and rotational positions were within a tolerance of 1 mm and 2° respectively. The translational and the rotational errors using this system were then documented.
The position verification by ExacTrac, Infrared Positioning System was followed by a set of images by Linac OBI (kv) system at 0° and 90° position. The images were autofused and compared with DRR generated from TPS. The extent of translational shift and rotational shift was then applied and documented.
The primary objectives were to quantify daily setup errors; the secondary objective was to use the data obtained to create a CTV–PTV margin for stereotactic fractionated radiotherapy. The systematic and random geometric uncertainties due to setup errors were calculated using alternate day, ExacTrac (kv), and OBI (kv) image data sets of 35 patients.
Systematic error (∑) is calculated as the average displacement of a particular reference structure and direction between simulation and treatment during the whole treatment course. Random component (σ) of any error is a deviation that can vary in direction and magnitude for each delivered treatment fraction. For each individual patient, random error is calculated as the dispersion around the systematic error. The mean errors represented the systematic errors, and the standard deviation of mean was considered as the random setup errors.
Statistical testing was conducted with the Statistical Package for Social Sciences System version SPSS 20.0 and R-3.2.0. Continuous variables were expressed as mean ± SD and were compared between methods of treatment of patients by using paired T-test. Categorical variables were presented as absolute frequencies with percentages. Systematic and random errors were calculated for each patient using the appropriate formula. All ‘P’ values will be two-tailed with significance defined as P < 0.05 at the level of 95% confidence limit. By incorporating the value of systematic and random error into Van Herk’s equation = PTV margin = 2.5 ∑ + 0.74 σ (minimum dose to CTV is 95% for 90% of the patients), PTV margin for each axis was calculated.
| Observations and Result|| |
A total of 35 patients were recruited prospectively for either adjuvant RT after tumor excision or definitive RT for brain tumors. The ages of the 15 female and 20 male patients were between 13 and 77 years with an average age of 54.5 years. The average irradiated PTV volume was 290.7 cm3, with a range of 2.4–912.5 cm3. Each patient received a dose of 50–60 Gy in 25–30 fractions with 6 MV beam energy. Correction and verification images were acquired on alternate day of treatment. A total of 2100 images were performed for the 35 patients on ExacTrac (kv) system and 1050 images on OBI (kv) system, and the data obtained in translational and rotational dimensions were used for analysis. [Figure 1] represents the sum of setup errors generated by ET and OBI in all dimensions whereas [Figure 2][Figure 3][Figure 4][Figure 5][Figure 6][Figure 7] represent the setup errors generated by ET and OBI in individual dimensions.
|Figure 1: Error generated by OBI (Kv) and ET (Kv) in different dimensions around isocenter|
Click here to view
|Figure 2: Setup errors around isocenter in Z-axis for OBI (Kv) and ET (Kv) imaging|
Click here to view
|Figure 3: Setup errors around isocenter in X-axis by OBI (Kv) and ET (Kv) imaging|
Click here to view
|Figure 4: Setup errors around isocenter in Y-axis by OBI (Kv) and ET(Kv) imaging systems|
Click here to view
|Figure 5: Setup errors around isocenter in yaw dimension by OBI (Kv) and ET (Kv) imaging|
Click here to view
|Figure 6: Rotational error in pitch dimension around isocenter for ET (Kv) imaging|
Click here to view
|Figure 7: Rotational errors around isocenter in roll dimension for ET (Kv) imaging|
Click here to view
Out of all 35 patients diagnosed with brain tumors, 9 patients had pituitary adenoma, 5 had meningioma, 2 patients had cranio-pharyngioma, 19 patients had glioma, 6 with GBM (glioblastoma multiforme) and 13 with other gliomas [Table 1]. This study has shown that the most common primary CNS tumor is gliomas, followed by pituitary adenoma and meningioma. McNeill in his study of epidemiology of brain tumors has reported same results.
The mean and random deviation of inter-fractional setup errors observed in this population study are shown in [Table 2].
CTV-PTV margins were calculated by incorporating the value of systematic and random errors into Van Herk’s equation =PTV margin = 2.5 ∑ + 0.74 σ (minimum dose to CTV is 95% for 90% of the patients), where ∑ is the systematic error and σ is the random error.
The CTV-PTV margin was calculated using data obtained from this study. The data obtained were put in Van Herk’s formula (PTV margin = 2.5 ∑ + 0.74 σ), and CTV-PTV margin was calculated.
CTV-PTV margin for ET was 2.70, 2.72, and 2.94 mm and 3.38o, 2.84o, 2.51o in lateral, longitudinal, vertical and pitch, roll, yaw dimensions, respectively. The CTV-PTV margin for OBI was 7.40, 6.20, 5.30 mm and 7.27o in lateral, longitudinal, vertical, and yaw dimensions, respectively [Table 3].
The overall CTV-PTV margin calculated for ET was 2.84 mm and for OBI was 6.5 mm.
| Discussion and Conclusion|| |
The aim of this study was to calculate daily setup errors and to determine an adequate safety margin (CTV-PTV) that would allow an acceptable dose distribution to target area with sharp dose fall to surrounding normal critical structures. In this study, the inter-fractional setup error was calculated and compared between daily orthogonal OBI (kv) and ExacTrac (kv) imaging in patients of brain tumor with Brain Lab mask on Novalis-Tx linear accelerator.
The setup errors were calculated for all 35 patients in 15–16 sessions on alternate day in order to avoid over exposure with radiation. Linthout et al. noted that the radiation dose per X-ray tube with the ExacTrac system is 0.5 mSv. For a single verification, the patient absorbs a dose of 1 mSv. This dose is very low compared with CBCT, where the dose is 14.0 mSv. This imaging dose of ET is negligible when compared with treatment dose and thus it can be ignored.
The population systematic errors with ExacTrac (kv) in translational component observed in this study were 0.83, 0.78, and 0.90 mm in the lateral (X-axis), longitudinal (Y-axis), and vertical (Z-axis) directions, respectively. The population systematic errors in the rotational component observed were 1.00o, 0.83o, and 0.72o in pitch, roll, and yaw directions, respectively [Table 4]. Population systematic errors calculated by Infusino et al., with ExacTrac in translational and rotational dimensions, were 1.21, 1.59, and 1.89 mm and 1.60o, 1.05o, 1.80o in lateral, longitudinal vertical and pitch, roll, and yaw dimensions, respectively. Se An Oh et al. calculated a margin of 0.72, 1.57, 0.97 and 0.72, 0.87, 0.83 in translational and rotational dimensions, respectively. Their findings of setup variation with ET are similar to the findings in our study.
In the present study, population systematic error with OBI (kv) observed was 2.24, 1.74, and 1.50 mm and 2.53o in lateral, longitudinal, vertical, and yaw directions, respectively [Table 4]. A study by Hong et al. calculated a setup variation with OBI imaging system, with BrainLAB mask stereotactic radiotherapy of 42 patients with 57 brain lesions. The mean and SD of the couch shift were 0.0±0.9, 0.1±1.4, and 0.3 ±0.8 mm in the vertical, longitudinal, and lateral directions, respectively. Systematic errors detected by Hong et al. is quite less (from 0 to 0.30 mm) compared to our result which is from 1.50 to 2.24 mm. Ali et al. in a study with OBI observed shift as large as 5.00, 8.00, and 4.50 mm in lat, long, and vertical dimensions.
The population random error with ExacTrac (kv) in translational components observed in this study was 0.90, 1.10, and 0.89 mm in lateral, longitudinal, and vertical dimensions, respectively. The population random errors in rotational components observed were 1.20o, 1.05o, and 0.96o, respectively. Infusino et al. evaluated random errors with ExacTrac as 0.20, 0.18, 0.27 and 0.25, 0.21, 0.17 in lateral, longitudinal, vertical, and pitch roll yaw dimensions, respectively. Se An Oh et al. observed errors of 0.31, 0.46, 0.54 and 0.28, 0.21, 0.17 in translational and rotational dimensions, respectively.
In the present study, the population random errors with OBI (kv) observed were 2.48, 2.53, 2.12 mm and 2.39° in lateral, longitudinal, vertical, and yaw dimensions, respectively, which is similar to the results observed by Ali et al., which was 2.20, 2.00, and 2.60 mm in lateral, longitudinal, and vertical dimensions.
CTV-PTV margin with ExacTrac (kv)
In this study, we have found a CTV-PTV margin in translational dimension, calculated by using Van Herk’s equation, which is 2.70, 2.72, and 2.94 mm in lateral, longitudinal, and vertical dimensions, respectively. CTV-PTV margin in rotational dimensions was 3.38o, 2.84o, and 2.51o in pitch, roll, and yaw dimensions, respectively. The overall CTV-PTV margin in the present study is 2.84 mm. In a study by Infusino et al., CTV-PTV margin was 0.5, 1.5, and 2.3 mm in lateral longitudinal and vertical dimensions, respectively. CTV-PTV margin in the rotational dimension was 1.5, 0.1, and 2.1, respectively. Se An Oh et al. reported a CTV-PTV margin of 0.97, 1.26, and 0.21 mm in lateral longitudinal and vertical dimensions, respectively. CTV-PTV margin in the rotational dimension was 1o, 0.7o, and 0.7o in pitch roll and yaw dimensions, respectively. Paul et al. recommended a PTV margin of 4.7, 5.1, and 4.3 in lateral, longitudinal, and vertical dimensions, respectively [Table 5]. Kumar et al. have recommended in their study using Gill–Thomas–Cosman relocatable frame in brain-fractionated stereotactic radiotherapy that 4 mm CTV-PTV margin is adequate for dose coverage of CTV. Drabik et al. in their study mentioned that in GBM, brain tumor patients’ margin in X, Y, and Z-directions is 3–4, 2, and 2–4 mm, respectively.
Beltran et al. reported inter- and intra-fractional positional uncertainties in pediatric brain tumor patients with megavoltage CBCT. The residual uncertainties were 0.5, 0.5, and 0.5 mm in the lateral, longitudinal, and vertical dimensions, respectively, for the systematic errors (Σ) and 0.9, 0.9, and 1.1 mm in the lateral, longitudinal, and vertical dimensions, respectively, for the random errors (σ). Weiss and Hess suggested that interobserver variability in target delineation was a major factor causing treatment inaccuracy. The delineation uncertainty can be reduced by using multimodality imaging as suggested by Rasch et al. The overall CTV-to-PTV margin, including tumor delineation, setup uncertainties, and internal organ displacement, should be fully investigated before determining PTV margin. In addition, IGRT minimizes the interfractional setup errors, and the margin derived from this study should not be used if IGRT is available.
CTV-PTV margin for OBI (kv):
In this study, CTV-PTV margin with Von Herk’s formula was 7.4, 6.2, 5.3 mm and 7.27o in lateral, longitudinal, vertical, and yaw dimensions, respectively. The overall CTV-PTV margin in the present study is 6.5 mm. Therefore, from the above discussion, we can infer that OBI-Kv imaging and setup correction system can be used as single imaging modality in brain tumor-fractionated radiotherapy at centers where ET is not available with some extra PTV margins around CTV.
Limitation of this study is small sample size, because of which the study is underpowered. Another limitation of our study is that we only considered bony rigid image registration when using the image fusion algorithms.
| Conclusion|| |
Daily setup verifications using the BrainLAB ExacTrac 6D image-guided system are very useful for evaluation and correction of the setup uncertainties while treating brain tumors. However, if we use OBI (Kv) orthogonal X-ray imaging system for setup verification and treatment of brain tumors, we have to consider an extended CTV-PTV margin as we have found in our study. Further studies with large sample sizes are required to achieve an adequate CTV-PTV margin while treating brain tumors using OBI-kv imaging system.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Shields LB, Coons JM, Dedich C, Ragains M, Scalf K, Vitaz TW, et al
. Improvement of therapeutic index for brain tumors with daily image guidance. Radiat Oncol 2013;8:283.
Ryu S, Khan M, Yin FF, Concus A, Ajlouni M, Benninger MS, et al
. Image-guided radiosurgery of head and neck cancers. Otolaryngol Head Neck Surg 2004;130:690-7.
Ryu S, Fang Yin F, Rock J, Zhu J, Chu A, Kagan E, et al
. Image-guided and intensity-modulated radiosurgery for patients with spinal metastasis. Cancer 2003;97:2013-8.
Infusino E, Trodella L, Ramella S, D’Angelillo RM, Greco C, Iurato A, et al
. Estimation of patient setup uncertainty using BrainLab ExacTrac X-ray 6D system in image-guided radiotherapy. J Appl Clin Med Phys 2015;16:5102.
Hurkmans CW, Remeijer P, Lebesque JV, Mijnheer BJ Set-up verification using portal imaging; review of current clinical practice. Radiother Oncol 2001;58:105-20.
Stroom JC, Heijmen BJ Geometrical uncertainties, radiotherapy planning margins, and the ICRU-62 report. Radiother Oncol 2002;64:75-83.
van Herk M, Remeijer P, Rasch C, Lebesque JV The probability of correct target dosage: Dose-population histograms for deriving treatment margins in radiotherapy. Int J Radiat Oncol Biol Phys 2000;47:1121-35.
McNeill KA Epidemiology of brain tumors. Neurol Clin 2016;34:981-98.
Oh SA, Yea JW, Kang MK, Park JW, Kim SK Analysis of the setup uncertainty and margin of the daily ExacTrac 6D image guide system for patients with brain tumors. PLoS One 2016;11:e0151709.
Hong L, Chen C, Garg M, Yaparpalvi R, Mah D Clinical experiences with onboard imager KV images for linear accelerator-based stereotactic radiosurgery and radiotherapy setup. Int J Radiat Oncol Biol Phys 2009;73:556-61.
Ali I, Tubbs J, Hibbitts K, Algan O, Thompson S, Herman T, et al
. Evaluation of the setup accuracy of a stereotactic radiotherapy head immobilization mask system using kv on-board imaging. J Appl Clin Med Phys 2010;11:3192.
Paul S, Roy S, Agrawal S, Munshi A, Jassal K, Ganesh T, et al
. Prospective comparative evaluation of planning target volume margin for brain intensity modulated radiotherapy utilizing hybrid online imaging modalities. Clin Cancer Investig J 2015;4:645. [Full text]
Kumar S, Burke K, Nalder C, Jarrett P, Mubata C, A’hern R, et al
. Treatment accuracy of fractionated stereotactic radiotherapy. Radiother Oncol 2005;74:53-9.
Drabik DM, MacKenzie MA, Fallone GB Quantifying appropriate PTV setup margins: Analysis of patient setup fidelity and intrafraction motion using post-treatment megavoltage computed tomography scans. Int J Radiat Oncol Biol Phys 2007;68:1222-8.
Beltran C, Krasin MJ, Merchant TE Inter- and intrafractional positional uncertainties in pediatric radiotherapy patients with brain and head and neck tumors. Int J Radiat Oncol Biol Phys 2011;79:1266-74.
Weiss E, Hess CF The impact of gross tumor volume (GTV) and clinical target volume (CTV) definition on the total accuracy in radiotherapy theoretical aspects and practical experiences. Strahlenther Onkol 2003;179:21-30.
Rasch C, Steenbakkers R, van Herk M Target definition in prostate, head, and neck. Semin Radiat Oncol 2005;15:136-45.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]